Equilibrium Bee Colony Algorithm for Global Optimization Problems

To overcome premature convergence of artificial bee colony algorithm, an equilibrium bee colony (EBC) algorithm was proposed for balance exploring ability between global search and local search. Onlookers were distributed randomly to employed bees, and then the employed bee's neighborhood was explored through differential mutation of the onlookers. For enhancing the population diversity information, a random individual group was generated; meanwhile a novel crossover operator was employed between onlookers and the random group. The experiments were carried out through 15 benchmarks, and the results showed that: when compared with artificial bee colony(ABC) algorithm and the self-adaptive differential evolution(SaDE), proposed in IEEE Transaction on evolutionary computation, the proposed algorithm performed better in term of efficiency and stability.